Work Environment and Job Demands Assessment for Teams and EmployersIt describes the job, not the jobholder. No total, and the page says why.
Thirty-eight original items about five conditions of a job — the amount of work, the say over it, the help when it sticks, how clearly it is spelled out and how much notice changes come with — drawn as a board between named ends, with the ordinary workplace marked. It measures a situation and never a person.
A conditions survey that refuses to become a verdict on anybody
The Work Environment and Job Demands Assessment is a thirty-eight item survey of five conditions of a job — quantitative demands, influence at work, social support, role clarity and predictability — each reported as a position between two named poles against an ordinary-workplace reference. It measures the job as described over a typical month, never the person describing it.
Most workplace surveys in this space slide from the job to the person within a page: a heavy load becomes a stress score, a thin support network becomes a resilience reading, and the respondent walks away with a label they did not sit down for. This one is built the other way round. Every item is written about the job — the work it is given, the say it allows, the help it provides, the expectations it states and the notice it gives — and the report opens, in the largest type on the page, with the sentence that it describes the job and not the jobholder.
The five dimension definitions follow the third version of the Copenhagen Psychosocial Questionnaire, which the COPSOQ International Network publishes openly and permits for non-commercial research. Every item here is an original work written to those definitions; no item text from that questionnaire or any other instrument is reproduced, and AssessAll is not affiliated with, endorsed by or licensed by the network or its authors.
There is no total, ever, and the report proves it rather than asserting it. Where a total would sit it prints two opposite jobs — a heavy load with little help, a light load with help close by — whose unweighted means are identical. More work is not the same kind of thing as more help, so their average is not a measurement of anything. The board is ordered by how far each condition sits from the ordinary mark in standard errors, never by which end of the bar looks nicer.
The ordinary mark is not the middle of the scale. Rating scales cluster above their midpoint, so a midpoint reference would flatter every workplace. Instead every item carries an authored share for each answer ordinary respondents give, the reference is computed from those, the shares are printed on the report, and observed data will replace them. Every bar carries its 68 and 95 per cent bands, a condition is called above or below the ordinary range only when its whole band clears the mark, and no condition carries a point figure or a percentile because none is built from eight items.
The page ends with the cost of the end of the bar the job sits at, one if-then change to the job rather than to the person, and the size of movement on a repeat sitting that would count as real change rather than measurement error.
What you walk away with
Work fits the hours, or work outruns the hours. Piling up, breaks cut short, deadlines arriving faster than work leaves.
Decided for the jobholder, or decided by the jobholder. Task order, method, amount, and which tasks are given.
Help is hard to get, or help is close at hand. Colleagues and the immediate supervisor, and whether they listen.
Expectations left unclear, or spelled out. Goals, who owns what, and what counts as finished.
Changes arrive unannounced, or come with notice. How far ahead, and whether the reasons come too.
What this page is not — a burnout, stress, wellbeing or mental-health reading, or a label on anybody — printed before any figure.
Inside your report
Illustrative sample — your report is generated from your own description of the job.
The filled block is the 68 per cent band and the outlined block the 95 per cent band; the shaded zone is the ordinary range. A condition is called above or below it only when its whole outlined block clears the mark. No point figure is printed.
| Two opposite jobs | Amount of work | Say over the work | Help when needed | Clear expectations | Notice of changes | Mean |
|---|---|---|---|---|---|---|
| Job A: heavy load, little help | 80 | 35 | 30 | 60 | 45 | 50 |
| Job B: light load, help close by | 20 | 65 | 70 | 40 | 55 | 50 |
The five conditions do not share a direction, so their mean is not a measurement of anything. Two opposite jobs share the same mean, which is why the report prints the refusal where a total would sit, and names no worst condition on level alone.
Built for
- Team leads and employers who want a plain reading of a job's conditions before a review, a redesign or a hiring round
- HR, people and works-council members who need a conditions survey that cannot be read as a diagnosis of anybody
- Occupational health and safety programmes that survey demands, support and clarity as properties of jobs, not of people
- Jobholders who want to describe their own job in five conditions and see where it sits against the ordinary workplace
Describe the job. See where it sits.
38 items across five formats · about 15 minutes · the conditions board, the ordinary mark, and one change to the job.
₹749 (incl. GST) · assessment and full report, nothing further to pay
Frequently asked questions
No. It measures five conditions of a job — the amount of work, the say over it, the help available, how clearly it is spelled out and how much notice changes come with — as you describe them over a typical month. It does not measure or estimate stress, burnout, wellbeing, resilience or mental health, it does not classify or label the respondent, and the report says so in its first block before any figure.
Because the five conditions do not share a direction. A heavy load with little help and a light load with help close by are opposite jobs with the same average, and the report prints both to show it. Each condition is reported on its own bar, and the board is ordered by how far each sits from the ordinary mark, not by level.
The position a job would have if it were described the way ordinary jobs are described. Every item carries an authored share for each answer, the mark is computed from those, the shares are printed on the report, and they will be replaced by observed data. It is deliberately not the middle of the scale, because rating scales cluster above their midpoint and a midpoint reference would flatter every workplace.
No, and it is built so that it cannot. No item asks about the person, no figure on the report is about the person, and nothing on it can be read as a health statement. It reads one job as one person described it; a team reading needs several people describing the same job.
About fifteen minutes for thirty-eight items. ₹749 in India, inclusive of GST, or US$7.99 elsewhere, one time, for the sitting and the full report. It is free to retake, and a repeat sitting on the same job three months later shows which conditions have moved beyond measurement error.
Each one takes a single capability, puts you inside the situations where it is actually tested, and scores your choices against published evidence — with a report designed for that capability alone, not a template. They span hiring, compliance, education, operations and personal skill.
Browse the catalogue →Methodology: Thirty-eight original items across five formats: sixteen six-point position items with a named pole at each end and no midpoint, eight balanced-keyed agreement statements on a six-point scale, five select-every-that-is-true checklists, five true-or-false descriptions and four match-the-following reading checks. The single declared response instruction is DESCRIPTIVE: describe this job as it has been over a typical month, as an observer of the job would - never how the respondent feels about it and never what the respondent would do. It is deliberately neither a knowledge instruction nor a behavioural-tendency instruction, because a conditions survey asks the respondent to report on a situation, and no item on the form mixes in either of the other two. Every item is written about the job - the amount of work it is given, the say it allows, the help it provides, the expectations it states and the notice it gives - and none is written about the person doing it. Construct statement: this measures five conditions of a job - quantitative demands, influence at work, social support from colleagues and supervisor, role clarity and predictability - each reported as a position between two named poles against an ordinary-workplace reference; it does not measure, estimate or imply the respondent's burnout, stress, wellbeing, resilience, mental health, personality, engagement or performance, it does not classify, score, rank or label the respondent, and nothing on the report is a health statement about anybody. The checklists and true-or-false items are descriptive, not tests: their key marks which statements describe the high pole of a dimension, and a ticked high-pole statement or an unticked low-pole statement moves the job's position toward that pole. The four matching exercises check that the survey's five words mean to the respondent what they mean to the survey; they carry no score, feed no figure, and are reported to the operator only as a note on how to read the board. Scoring is a bipolar conditions profile with a refused composite. Each dimension is the mean position of its answered items on a 0-100 scale between its poles, with reverse-keyed items reflected; an unanswered item leaves the denominator and its reference term together. The reference for each dimension is not the scale midpoint but the DECLARED MARGINAL: every item carries an authored distribution of the answers ordinary respondents give, and the reference is the position a job would have if it were described the way ordinary jobs are described. Rating scales cluster above their midpoint, so a midpoint reference would flatter every workplace; the declared marginals are printed on the report and will be replaced by observed marginals once live data exists. Reliability is McDonald's omega estimated from the number of answered items and a stated assumed inter-item correlation of .32; the standard error of each dimension is its declared-marginal standard deviation times the square root of one minus omega, and every dimension carries its 68 and 95 per cent bands. A dimension is called above the ordinary range only when its whole 95 per cent band lies above the reference, below only when the whole band lies below it, and inside otherwise. No dimension here is built from eight or more items, so no dimension carries a point figure or a percentile: each prints a placement, a drawn band and its band edges in words, and the refusal of a figure is printed in the reliability table rather than performed silently. Dimensions are ordered by the size of their departure from the reference in standard errors, never by level, because a low reading on help and a high reading on demands are not the same kind of thing and neither is a worst. There is no composite, ever: the five dimensions do not share a direction, so their mean is meaningless, and the report demonstrates this by printing two opposite authored profiles whose unweighted means are identical. A careless-responding composite and the consistency of one deliberately paired item are computed for the operator only and never shown to the respondent as a judgement. Construct spine and sources: the dimension definitions follow the Copenhagen Psychosocial Questionnaire, third version (Burr, Berthelsen, Moncada, Nubling, Dupret, Bocereau, Vleeschouwer, Wolf, Ptacek and others, 2019, Safety and Health at Work) and the guidelines of the COPSOQ International Network, which publishes its dimensions and their definitions openly and permits non-commercial research use; the first and second versions of the questionnaire (Kristensen, Hannerz, Hogh and Borg, 2005, Scandinavian Journal of Work, Environment and Health; Pejtersen, Kristensen, Borg and Bjorner, 2010, Scandinavian Journal of Public Health); the German validation of the instrument (Nubling, Stossel, Hasselhorn, Michaelis and Hofmann, 2006, GMS Psycho-Social-Medicine); the job demands and decision latitude model (Karasek, 1979, Administrative Science Quarterly; Karasek and Theorell, 1990, Healthy Work) and its extension with social support (Johnson and Hall, 1988, American Journal of Public Health); the job demands-resources model (Bakker and Demerouti, 2007, Journal of Managerial Psychology; Demerouti, Bakker, Nachreiner and Schaufeli, 2001, Journal of Applied Psychology); the role ambiguity construct (Rizzo, House and Lirtzman, 1970, Administrative Science Quarterly); the treatment of aggregated perceptions of a shared situation (Bliese, 2000, in Klein and Kozlowski, Multilevel Theory, Research and Methods in Organizations); the sources on response style and balanced keying (Baumgartner and Steenkamp, 2001, Journal of Marketing Research) and on common method variance (Podsakoff, MacKenzie, Lee and Podsakoff, 2003, Journal of Applied Psychology); the item-writing guidelines of Haladyna, Downing and Rodriguez (2002, Applied Measurement in Education); and McDonald (1999, Test Theory: A Unified Treatment) for omega. Every item in this instrument is an original work written for AssessAll: no item text from the Copenhagen Psychosocial Questionnaire or any other instrument is reproduced, the survey uses only the published dimension definitions as its construct spine, and AssessAll is not affiliated with, endorsed by or licensed by the COPSOQ International Network, any of the authors named above, or the owner of any other instrument or certification.